[evlatests] Antenna 16, Q-band, bad polarizer?

[Rick Perley]

    More interesting -- or at least curious -- information from the 
short test today. 

    In plotting the visibility amplitudes at Q-band, I found large 
'closure errors', of the order of a few percent.  A quick check showed 
that *all* of these come from antenna 16.  I suspected bad delay or 
bandpass (since I created a 100 MHz-wide continuum by integrating across 
the spectrum) -- but no, that wasn't the problem, since the closure 
errors remained after proper delay and bandpass calibration.
    The problem is a bad polarizer.  In plotting the cross-polarization 
amplitudes, I find that this antenna shows a 10% cross-polarization!  I 
guessing that this is cross-coupling the correlations, either through 
the D*P  mechanism, or the D*D mechanism.  Neither is corrected with the 
current way we handle polarization calibration ...